In vitro and in vivo evaluation of quinones from Auxemma oncocalyx Taub. on Leishmania braziliensis

Departamento de Patologia e Medicina Legal, Faculdade de Medicina, Universidade Federal do Ceará (UFC), Rua Alexandre Baraúna, 949, 60.430-160 Fortaleza, CE, Brazil. Departamento de Farmácia, Faculdade de Farmácia, Odontologia e Enfermagem, UFC, Rua Capitão Francisco Pedro 1210, 60430-370 Fortaleza, CE, Brazil. Departamento de Fisiologia e Farmacologia, Faculdade de Medicina, UFC, Av. Cel. Nunes de Melo, 1127, 60430-270 Fortaleza, CE, Brazil.


INTRODUCTION
Leishmaniasis are protozoan diseases which represent a risk for 350 million people worldwide, and 2 million new cases occur yearly (World Health Organization [WHO], 2010).Etiologic agents are intracellular parasites of the genus Leishmania that display a spectrum of a manifestation which goes from cutaneous involvement ith late destruction of mucous membranes to generalized systemic visceral disease with fatal outcome, if not treated (Pearson et al., 2000;Desjeux, 2004).There is still no effective vaccine to control the wide range of disease caused by different species of Leishmania (Oliveira et al., 2009).The disease may regress spontaneously or evolve, thus requiring treatment (Piscopo and Mallia, 2006).The antileishmanial first-line drugs are the pentavalent antimonials, meglumine antimoniate (Glucantime) and sodium stibogluconate (Pentostam).They are generally effective during the acute infection stages but not against the late stages, and produce significant side effects due to high toxicity and tissue drug accumulation, which includes myalgias, nausea, vomiting, cardiac arrhythmia, hepatitis, or pancreatitis (Croft et al., 2006;Palumbo, 2009).Amphotericin B and pentamidine are better tolerated drugs, but require long courses of parenteral administration (Amato et al., 2008).Miltefosine and fluconazole have recently showed effectiveness against cutaneous leishmaniasis (CL) caused by Leishmania braziliensis (Machado et al., 2010;Sousa et al., 2011), but despite the lower toxicity, these second line drugs are not useful against other forms of leishmaniasis (Palumbo, 2009).Furthermore, the continuous use of ineffective drugs has led to the development of resistance to their compounds (Escobar et al., 2001), which have stirred an urgent need for novel, effective, and safe drugs for treatment of leishmaniasis.
Since plant derivatives are among the most active agents against infections (Delorenzi et al., 2001), many researchers have been looking for better effects and less toxicity to treat leishmaniasis in these substances, especially those already used by people living in endemic areas (Weniger et al., 2001).Auxemma oncocalyx Taub.belongs to the Boraginaceae family, it is known as "pau branco" (white wood), and is easily available by rural settlers in Northeastern Brazil.The stem bark of the tree is astringent and popularly used in the treatment of wounds (Braga, 1976).
Previous pharmacological studies reported that the hydroalcoholic extract of the stem presents antioxidant, analgesic and anti-inflammatory properties (Ferreira et al., 2004(Ferreira et al., , 2008)).Alantoin and -sitosterol glycoside have been isolated from A. oncocalyx and are responsible for some of its pharmacological properties (Pessoa and De Lemos, 1997).Besides these compounds, at least six quinones were successfully isolated and oncocalyxone A, obtained in significant amounts, seems to be partially involved in the bioactivity of the plant (Leyva et al., 2000).Several quinones present antileishmanial (Sauvain et al., 1993;Sittie et al., 1999;Teixeira et al., 2001), antitumoral (Morello et al., 1995;Itoigawa et al., 2003), antifungal (Perry et al., 1991;Gafner et al., 1996), or antimalarial activity (Figueiredo et al., 1998), either in vitro or in vivo.The aim of the present work was to evaluate the in vitro and in vivo activity of the quinone fraction from A. oncocalyx Taub.against L. braziliensis.

Plant extraction and purification of the quinone fraction
A. oncocalyx Taub.was collected in the city of Pentecoste, Ceará  (Pessoa et al., 1993).

Parasites and animals
Three  Oliveira et al., 2004).The parasites, stored in liquid nitrogen, were thawed and cultured as promastigotes at 26°C in Schneider's insect medium (Sigma-Aldrich, Chemical Co., St. Louis, USA) supplemented with 10% heat-inactivated fetal calf serum (Sigma), 2% sterile normal human urine, 2 mM L-glutamine (Gibco BRL, Grand Island, NY), and antibiotics [100 U/ml penicillin, 100 µg/ml streptomycin sulfate (Sigma-Aldrich)].Subcultures were made in the stationary phase of growth and parasites were used at no later than the fourth passage.Prior to infection, promastigotes were harvested from culture, washed in sterile saline, counted in Neubauer's chamber and adjusted to the appropriate concentration.

Anti-promastigote activity
For the tests in vitro against promastigotes in 96-well plates stationary-phase promastigotes were added at a concentration of 10 6 cells/well in Schneider medium supplemented with 10% heatinactivated fetal calf serum (Sigma), 2% sterile normal human urine, 2 mM L-glutamine (Gibco), and 100 U/ml penicillin + 100 µg/ml streptomycin sulfate (Sigma-Aldrich) determined after counting in a Neubauer chamber.Drugs were diluted with DMSO 5% and placed at concentrations of 0.01 to 10 μg/mL.Amphotericin B was chosen as the control drug.The plates were incubated in a biochemical oxygen demand (BOD) at 24°C, and after that, 1 µCi per well of [3H]thymidine (Amersham International, Amersham, UK) was added and the cells were incubated for another 24 h and harvested.
[3H]thymidine incorporation was measured in a β-counter (Pharmacia, Finland) after washing to distinguish the non-used thymidine from the one incorporated in DNA.Assays were done in duplicate and made a replica of each test.The inhibition of growth was expressed as the percent decrease of radioactive incorporation in treated cells when compared with untreated control.It was considered a good antileishmanial activity when values equal or greater than 70% were obtained.The formula for calculation of antileishmanial activity was:

Infection, treatment and lesion development
Hamsters were infected subcutaneously in the right hind footpad with 10 6 stationary phase L. braziliensis promastigotes in 20 µl of sterile saline.Treatment was initiated two weeks after inoculation when lesions were well defined.Animals were randomly divided into groups of six to eight and the drugs administered daily for 28 consecutive days.QF was dissolved in distilled water containing 0.1% of Tween 80 and 0.5% of carboxymethylcellulose (CMC) and administered using the following routes: (a) oral (p.o.), with 10 or 20 mg/kg by intragastric intubation; (b) intraperitoneal (i.p.) with 20 mg/kg.Alternatively, Glucantime (Sanofi-Aventis Farmacêutica, São Paulo, Brazil) was injected at the dose of 60 mg/kg/day intramuscularly (i.m.).Control groups received p.o. or i.p. drugless vehicle (Tween 80 + CMC + H 2 O) at equivalent volumes.Before treatment was conducted, a toxicity test using various concentrations of QF that demonstrated concentrations of 10 and 20 mg/kg did not kill the animals.Lesion sizes were measured weekly with a dial gauge caliper (Mitutoyo, 0.01 mm sensitivity) and expressed as the difference between the thicknesses (mm) of the infected and contralateral uninfected footpads.

Treatment outcome
The number of parasites in the popliteal lymph was quantified by the limiting dilution technique as previously described (Lima et al., 1997).Briefly, after the treatment, the animals were euthanized by inhalation of halothane (Sigma-Aldrich) and submerged in 3% iodized alcohol up to 3 min to allow decontamination.The lymph nodes were removed aseptically and macerated in a Petri dish with 2 ml of Schneider medium.After removal of debris by sedimentation for 5 min, the homogenates were serially diluted (1:10) in Schneider's medium supplemented with 100 U/ml of penicillin/ml, 100 μg/ml of streptomycin/ml, 10% fetal calf serum and 2% sterile human urine.One hundred microliters of these dilutions was distributed into 96-well flat bottom plates containing agar-blood in 6 replicates per concentration.The plates were incubated at 25°C and observed under an inverted microscope (Nikkon, Japan) every 3 days, up to a maximum of 30 days, to record the dilutions containing promastigotes.The final number of parasites per tissue was determined using the ELIDA software, version 12c for window (Taswell, 1984).

Statistical analysis
The data are presented as mean ± standard error of the mean.The significance of the results relating to the parasite load was calculated by Mann-Whitney test.The anti-promastigota activity and lesion sizes from treated and untreated animals were analyzed by the one-way analysis of variance (ANOVA) and complemented by the Bonferroni test for multiple comparisons.All analysis and graphs were performed using GraphPad Prism version 5.0 (GraphPad Software, San Diego, USA).Values of P < 0.05 were considered significant.

RESULTS
QF displayed marked in vitro anti-promastigote activity at concentration of 1 and 10 μg/mL (80.5 ± 2.80 and 91.0 ± 0.65%, respectively) against L. braziliensis, being similar to Amphotericin B (Figure 1).To determine the effect of QF on L. braziliensis infection in hamsters, the animals days via oral.The results showed that hamsters treated with QF (10 or 20 mg/kg, p.o.) did not present significant reduction of the lesion thickness as compared to untreated animals.Also, doses of 20 and 10 mg/kg (p.o.), did not produce significant reduction of parasite load in lymph nodes as compared to untreated group (Figure 2A  and B).As expected, animals treated with glucantime showed antimonial drug effectively brought footpad sizes down to normal within 2 weeks of treatment and also suppressed parasite growth in lymph nodes (Figure 2A  and B).
To investigate if a different administration route of the drug could be more effective on controlling L. braziliensis infection, hamsters were treated with 20 mg/kg QF intraperitoneally for 28 days (the highest dose, but yet not toxic to animals).The results showed that the lesion size of hamsters treated with QF i.p. decreased by 65% (P < 0.05), and parasite burden in the lymph nodes was significantly lower (9.3 ± 3.5 × 10 5 ) when compared with untreated controls (1.98 ± 2.0 × 10 6 ) (Figure 3A and B).

DISCUSSION
Antileishmanial activity has been reported in several compounds extracted from medicinal plants belonging to diverse chemical groups, including quinones (Chan-Bacab et al., 2001;Rocha et al., 2005).Other studies have found simple quinones isolated from dried trunks of Jacaranda copaia to present significant anti-promastigote and anti-amastigote activities in vitro against L. amazonensis, but only weak activity when tested against L. amazonensis-induced lesion in mice (Sauvain et al., 1993).Benzoquinones were found to be active in vitro against trypanosomes, but this has not been confirmed in vivo (Grady et al., 1984;Pahn et al., 1988).
In murine model, Leishmania major-and Leishmania donovani-infected BALB/c treated with buparvaquone formulation (BPQ), showed parasite burden decrease in lesions and liver, smaller and not ulcerated lesions, in comparison with untreated control (Garnier et al., 2007).Recently, a study using the molecular hybridization of a naphthoquinone core with a pterocarpan moiety (LGB-118) led to significant reduction in skin lesions development, swelling, ulceration and parasite burden of BALB/c L. amazonensis-infected (Da Cunha-Júnior et al., 2011).Also, acetylisolapachol, a hydroxyquinone derivative, showed in vitro activity against L. braziliensis, and in vitro and in vivo against L. amazonensis (Lima et al., 2004).
The leishmanicidal activity of a drug may be selective and direct against the parasite, or it may act indirectly by activating macrophage microbicidal mechanisms for instance.According to in vitro model systems, the macrophage microbicidal response to Leishmania infection can follow two distinct pathways.Upon infection, promastigotes elicit a respiratory burst with the generation of reactive oxygen intermediates such as hydrogen peroxide (H 2 O 2 ), • OH radical, superoxide (  and peroxynitrate as part of an oxygen dependent mechanism to kill promastigotes, however, a small percentage of phagocytosed organisms can survive (Beaman and Beaman, 1984).Second, murine or human macrophages can be activated to kill intracellular amastigotes, the form present during established infection, by previous exposure to cytokines such as IFN- and TNF-, which activate both oxidative and non-H 2 O 2 -associated microbicidal mechanisms (Bogdan et al., 1990;Liew, 1992;Assreuy et al., 1994;McSorley et al., 1996;Panaro et al., 1999).Studies have demonstrated that both H 2 O 2 -associated and non-H 2 O 2associated pathways contribute to Leishmania killing and that their relative degree of importance may differ during initial promastigote invasion versus established amastigote infection (Chang, 1983;Murray and Nathan, 1999;Erel et al., 1999).Quinones are highly redox active molecules and with their semiquinones radicals can lead to formation of reactive oxygen species (ROS), including • O 2  , H 2 O 2 , and ultimately to hydroxyl radicals (Bolton et al., 2000).The formation of ROS could probably explain the in vitro leishmanicidal activity of the quinones in this study, since promastigotes are readily susceptible to killing by H 2 O 2 in vitro (Murray, 1981;Zarley et al., 1991).It has been shown that Leishmania chagasi promastigotes in vitro are susceptible to killing by both H 2 O 2 and the redox-cycling compound menadione, a quinone that causes the generation of • O 2  in the presence of promastigotes (Wilson et al., 1994).However, this source of quinone free radicals seems to be often more readily apparent in vitro than in vivo.Menadione providing an excellent example of this phenomenon, although presents anticancer activity in vitro in combination with other chemotherapeutic agents.However, menadione does not demonstrate the same activity in vivo even at high doses (Nestor et al., 1991;Djuric et al., 1995).Besides, in a BALB/c mouse model of leishmaniasis, sublethal concentrations of menadione caused L. chagasi promastigotes to become more virulent (Wilson et al., 1994).
When QF was administered intraperitoneally at a dose of 20 mg/kg, the lesion size was reduced by 65%, but it was however not able to promote parasite killing as evidenced by the mean number of parasites in the lymph nodes as compared to controls.Reduction in the lesion size induced by Leishmania infection can not necessarily mean a decrease in the parasite load, but only a decrease of the local inflammatory reaction.QF of A. oncocalyx has a wide range of biological effects including anti-parasitic, antitumoral and antiplatelet activities (Leyva et al., 2000;Ferreira et al., 2008).It have also been shown that the compound presents anti-inflammatory and antiedematogenic activities, reducing the effect of carrageenan (Ferreira et al., 2004).Therefore, it is suggested that the reduction in the lesion size produced by 20mg/kg, i.p of QF in L. braziliensis-infected hamsters may probably be due to its anti-inflammatory effect.In addition, the observation that QF when used via oral did not offer significant protection in hamsters infected by L. braziliensis, suggests that QF can possibly be converted into non-active metabolite(s) or otherwise be inactivated either by reduction or by interaction with Med.Plants Res.serum proteins.Quinones can be metabolized by various routes: substitution or reductive addition with nucleophilic compounds or one and two-electron reductions (Koster, 1991).Driscoll et al. (1974) found that the biological activity of some quinones, as lapachol and its analogs, is directly related to their chemical structures, thus any structural alteration in vivo will result in an inactive product or will abolish their biological activities (Teixeira et al., 2001).
Despite QF not to have demonstrated an anti-parasitic effect in L. braziliensis-infected hamsters, the production of quinones derivatives might insure their interest as antileishmanial candidate drugs.Furthermore, alternative therapy derived from medicinal plants opens new perspectives towards the development of effective, readily available and less-costly drugs for the treatment of the leishmaniasis in endemic areas.

Figure 1 .
Figure 1.Antileishmanial activity in vitro of a quinone fraction (QF) isolated from the heartwood of Auxemma oncocalyx Taub.against promastigores of L. braziliensis.The inhibition of growth was expressed as the percent decrease of radioactive incorporation in treated parasites when compared with untreated control.***P < 0.05 (test ANOVA).

Figure 2 .
Figure 2. Effect of oral treatment with QF (20 or 10 mg/kg).(A) Lesion growth, and (B) Parasite load in lymph node.Hamsters were infected with 10 6 L.braziliensis promastigotes (8 per group).Animals were left untreated or were treated with daily dose of 20 mg/kg QF or with 10 mg/kg or 60 mg/kg/i.m. glucantime.Treatment started three weeks post infection.Lesion thickness was measured weekly using a dial gauge caliper (mean  SE, n = 8).Parasite load was evaluated after the treatment.*P < 0.05 (A, Glucantime versus untreated or QF treatment; B, Glucantime versus QF treatment).

Figure 3 .
Figure 3.Effect of intraperitoneal treatment with QF (20 mg/kg).(A) Lesion growth, and (B) Parasite load in lymph node.Hamsters were infected with 10 6 L. braziliensis promastigotes (6 per group).Animals were left untreated or were treated with daily dose of 20 mg/kg QF or 60 mg/kg/i.m. glucantime.Treatment started three weeks post infection.Lesion thickness was measured weekly using a dial gauge caliper (mean  SE, n = 8).Parasite load was evaluated after the treatment.*P < 0.05 (A, QF treatment versus untreated; B, QF treatment versus untreated or Glucantime).